Road signs vary in size, shape, images, and words to communicate to a driver. For example, many signs have similar perimeter shapes yet convey different messages or images. Images on a sign, like an arrow, can convey different meanings depending on the other features on the sign. For example, a sign with an arrow might indicate a “one way” road, while another similar-sized and colored sign with an arrow might indicate the direction to a destination. Critical road signs, whose function is to provide important information for safe vehicle operation, can be difficult to identify and differentiate from the broader range of signs with similar shapes.
The electronic and communication capabilities available could aid in identification and communication between the road sign and a vehicle. Camera-based vision systems in particular are becoming common in new vehicles. However, current vision systems typically have difficulty accurately reading the broad range of highway road signs. For example, as discussed, a reader might identify the arrow, but not accurately distinguish the context of identifying whether the arrow might indicate a “one way” sign or a direction to a destination. There is a need for an improved identification system to more accurately detect or recognize, or both detect and recognize critical road signs, roadway information, and adjacent roadway information.